Cross-basin hierarchical scheduling method based on runoff space-time complementary characteristic evaluation

Abstract

The invention discloses a cross-basin hierarchical scheduling method based on runoff space-time complementary characteristic evaluation. According to the method, a complete and quantifiable index system is constructed, and the natural phenomenon of'runoff space-time complementation 'is converted into accurate mathematical representation. Therefore, an objective standard is provided for the judgment of strong complementation, and the scientificity of decision making is remarkably improved. In addition, the cross-basin runoff complementation problem is systematically analyzed from the two dimensions of consistency and difference for the first time. The method not only pays attention to runoff characteristics of the drainage basins, but also focuses on depicting interaction relations (such as high and low encountering and fluctuation counteracting) among the drainage basins, and provides a systematic tool for comprehensively evaluating the cross-drainage-basin complementary potential.

Description

Technical Field

[0001] This invention relates to the technical field of water resource scheduling decision-making, and specifically to a cross-basin hierarchical scheduling method based on the assessment of runoff spatiotemporal complementary characteristics. Background Technology

[0002] With the intensification of global climate change and the increasing frequency of extreme weather events (such as severe droughts, persistent torrential rains, and extreme heat), the safety and stable supply of water resources in river basins pose a severe challenge. Regional power grids, which rely primarily on hydropower, as well as agricultural irrigation and urban water supply systems that depend on stable water sources, face significant risks under extreme weather conditions. A single river basin may experience prolonged droughts, leading to a sharp decline in power generation and water shortages; or prolonged abundant water, resulting in water wastage or even flooding risks. Inter-basin water allocation is a key means of addressing this challenge. By constructing interconnection projects, surplus water resources from one river basin can be allocated to water-scarce basins, achieving peak shaving and valley filling, and enhancing the overall resilience and reliability of the water resource system.

[0003] The key to cross-basin scheduling is accurately identifying the spatiotemporal complementarity of runoff between different basins. However, existing research has significant shortcomings, mainly in the following aspects: Existing methods often focus on runoff characteristic analysis at single-basin stations or only conduct simple statistical correlation analyses between pairs of basins, lacking a comprehensive measure of the overall complementarity of complex systems composed of multiple basins. A quantitative indicator system capable of systematically and comprehensively characterizing the temporal and spatial complementarity of cross-basin runoff has not been established. Existing research often relies on single indicators (such as correlation coefficients), which are insufficient to reflect the multidimensional characteristics of complementarity, such as consistency and runoff fluctuation comparisons. Furthermore, at the level of scheduling decision support, there is a lack of clear and operational quantitative criteria and grading thresholds for defining "strong complementarity" and "weak complementarity." This leads to a high degree of subjectivity in assessing cross-basin complementarity potential, and more importantly, it fails to directly translate the complementarity assessment conclusions into differentiated scheduling strategies at different levels. Existing research is insufficient to support refined scheduling decisions regarding "when to adjust, how much to adjust, and how to adjust," resulting in cross-basin scheduling often remaining at the level of principle guidance. This fails to achieve precise and optimized coordinated allocation of water resources, thus hindering the full realization of the benefits of interconnection projects.

[0004] In summary, among the relevant technologies, there are technical problems such as one-sided evaluation methods for cross-basin runoff complementarity characteristics, and the evaluation results are highly subjective, which makes it difficult to support precise water resource allocation decisions. Summary of the Invention

[0005] The technical problem this invention aims to solve is that cross-basin runoff complementarity assessment methods are one-sided and the assessment results are highly subjective, thus making it difficult to support precise water resource allocation decisions. The purpose is to provide a cross-basin hierarchical allocation method based on runoff spatiotemporal complementarity assessment. This solves the technical problem that cross-basin runoff complementarity assessment methods are one-sided and the assessment results are highly subjective, thus making it difficult to support precise water resource allocation decisions.

[0006] This invention is achieved through the following technical solution:

[0007] In a first aspect, the present invention provides a cross-basin hierarchical scheduling method based on the assessment of runoff spatiotemporal complementarity characteristics, the method comprising:

[0008] Acquire runoff time series data from multiple watersheds within a preset scheduling period;

[0009] Based on the runoff time series data, time dimension indicators and spatial dimension indicators are calculated; wherein, the time dimension indicators are used to assess the temporal synchronicity of runoff changes between watersheds, and the spatial dimension indicators are used to assess the differences in runoff volume surplus and deficit and fluctuation between watersheds.

[0010] Based on the preset entropy weight method, the objective weights corresponding to each indicator are determined according to the numerical distribution of the time dimension indicators and the spatial dimension indicators.

[0011] The standardized index values ​​are weighted and summed according to the objective weights to generate a spatiotemporal comprehensive complementarity index.

[0012] The spatiotemporal comprehensive complementarity index is converted into a Z-index that follows a standard normal distribution;

[0013] Based on the quantile interval of the Z-index under the standard normal distribution, the runoff complementarity level among multiple watersheds is determined.

[0014] Based on the runoff complementarity level, scheduling control signals are generated for controlling inter-basin water transfer projects.

[0015] Furthermore, the time dimension indicators include a first time dimension indicator and a second time dimension indicator; wherein, the first time dimension indicator is a consistency indicator for assessing the consistency of runoff abundance and scarcity among watersheds, and the second time dimension indicator is a consistency indicator for assessing the consistency of runoff change trends among watersheds.

[0016] The spatial dimension indicators include a first spatial dimension indicator and a second spatial dimension indicator; wherein, the first spatial dimension indicator is a surplus difference indicator used to assess the runoff surplus compensation capacity between watersheds, and the second spatial dimension indicator is a fluctuation difference indicator used to assess the degree of runoff fluctuation offsetting between watersheds.

[0017] Furthermore, the step of calculating the time dimension index and the spatial dimension index based on the runoff time series data includes: the step of calculating the first time dimension index and the step of calculating the second time dimension index;

[0018] The steps for calculating the first time dimension index include:

[0019] Based on a preset runoff cumulative frequency threshold, the runoff data for each time period are divided into multiple abundant and scarce levels.

[0020] Based on the abundance and scarcity levels of runoff sequences in two watersheds at various time periods, the abundance and scarcity consistency index is calculated according to the number of time periods with the same level, the number of time periods with a level difference of the first preset level, the number of time periods with a level difference of the second preset level, and the number of time periods with opposing levels.

[0021] The steps for calculating the second time dimension index include:

[0022] Calculate the rate of change of the runoff time series data in adjacent time periods;

[0023] Based on the difference in the runoff change rate between the two watersheds at corresponding time periods, a consistency index of change trend is calculated.

[0024] Furthermore, the step of calculating the time dimension index and the spatial dimension index based on the runoff time series data includes: the step of calculating the first spatial dimension index and the step of calculating the second spatial dimension index;

[0025] The step of calculating the first spatial dimension index includes:

[0026] Based on the preset multi-year average runoff of each watershed, the runoff deviation value in each time period is calculated.

[0027] For each time period, if the runoff deviation values ​​of the two watersheds have opposite signs, the compensable water volume for that time period is determined based on the absolute value of the runoff deviation values ​​of the two watersheds.

[0028] The surplus difference index is calculated based on the ratio of the sum of compensable water volume for all time periods during the scheduling period to the sum of the total runoff of the two watersheds.

[0029] The steps for calculating the second spatial dimension index include:

[0030] Calculate the runoff changes in the multiple watersheds during each time period within the scheduling period;

[0031] The fluctuation difference index is calculated based on the algebraic sum of the runoff changes in the multiple watersheds during the same period.

[0032] Furthermore, the step of determining the objective weights corresponding to each indicator based on the preset entropy weight method and the numerical distribution of the time dimension indicators and the spatial dimension indicators includes:

[0033] The consistency index between abundant and scarce periods, the consistency index of changing trends, the surplus difference index, and the volatility difference index are standardized; wherein, the surplus difference index is standardized using a positive standardization method, and the consistency index between abundant and scarce periods and the volatility difference index are standardized using a negative standardization method.

[0034] Based on the standardized indicator values, calculate the information entropy of each indicator;

[0035] Calculate the corresponding objective weight based on the information entropy of each indicator; the greater the dispersion of the indicator, the smaller its information entropy, and the greater the weight assigned to it.

[0036] Further, the step of converting the spatiotemporal comprehensive complementarity index into a Z-index that follows a standard normal distribution includes:

[0037] The skewness coefficient of the spatiotemporal comprehensive complementarity index is calculated to quantify the asymmetry of its distribution;

[0038] Based on the skewness coefficient, the spatiotemporal comprehensive complementarity index is normalized to generate a Z-index that follows or approximately follows a standard normal distribution; wherein, the Z-index is used to eliminate the influence of the spatiotemporal comprehensive complementarity index on the complementarity level determination caused by the difference in the original distribution form.

[0039] Further, the step of determining the runoff complementarity level among multiple watersheds based on the quantile interval of the Z-index under the standard normal distribution includes:

[0040] The Z-index is compared with a plurality of preset thresholds; wherein the plurality of thresholds include a first preset threshold, a second preset threshold, a third preset threshold and a fourth preset threshold whose values ​​decrease sequentially, and the thresholds are determined based on the quantiles of a standard normal distribution.

[0041] Based on the comparison results, the runoff complementarity levels are divided into five levels: extremely strong complementarity, strong complementarity, average complementarity, weak complementarity, and extremely weak complementarity; among them...

[0042] When the value of the Z-index is greater than or equal to the first preset threshold, the complementarity level is determined to be extremely strong complementarity.

[0043] When the value of the Z-index is less than the first preset threshold and greater than or equal to the second preset threshold, the complementarity level is determined to be strong complementarity.

[0044] When the value of the Z-index is less than the second preset threshold and greater than or equal to the third preset threshold, the complementarity level is determined to be flat complementarity.

[0045] When the value of the Z-index is less than the third preset threshold and greater than or equal to the fourth preset threshold, the complementarity level is determined to be weak complementarity.

[0046] When the value of the Z-index is less than the fourth preset threshold, the complementarity level is determined to be very weak complementarity.

[0047] Secondly, the present invention provides a cross-basin hierarchical scheduling device based on runoff spatiotemporal complementarity assessment, comprising:

[0048] The acquisition module is used to acquire runoff time series data of multiple watersheds within a preset scheduling period;

[0049] The calculation module is used to calculate time dimension indicators and spatial dimension indicators based on the runoff time series data; wherein, the time dimension indicators are used to assess the temporal synchronicity of runoff changes between watersheds, and the spatial dimension indicators are used to assess the differences in runoff volume surplus and fluctuation between watersheds.

[0050] The determination module is used to determine the objective weights corresponding to each indicator based on the numerical distribution of the time dimension indicators and the spatial dimension indicators, according to a preset entropy weight method.

[0051] The generation module is used to generate a spatiotemporal comprehensive complementarity index by weighted summation of the standardized index values ​​according to the objective weights.

[0052] A conversion module is used to convert the spatiotemporal comprehensive complementarity index into a Z-index that follows a standard normal distribution;

[0053] The output module is used to determine the runoff complementarity level between multiple watersheds based on the quantile interval of the Z-index under the standard normal distribution.

[0054] The generation module is used to generate scheduling control signals for inter-basin water transfer projects based on runoff complementarity levels.

[0055] Thirdly, the present invention provides an electronic device, comprising: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, the instructions being executed by the one or more processors to cause the one or more processors to implement the method described above.

[0056] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0057] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0058] This invention constructs a quantitative index system with time and space dimensions, uses the entropy weight method to objectively assign weights and synthesize a comprehensive index, and then performs normalization transformation and standard classification. This achieves a technological leap from subjective experience judgment to objective machine evaluation of cross-basin runoff complementarity characteristics, providing standard and reliable technical parameter inputs for core water resource scheduling. Attached Figure Description

[0059] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0060] Figure 1 A flowchart illustrating a cross-basin hierarchical scheduling method based on runoff spatiotemporal complementarity assessment, provided in the embodiments of this specification;

[0061] Figure 2 This diagram illustrates the kernel density estimation and the identification of runoff abundance / drainage based on cumulative frequency provided in the embodiments of this specification.

[0062] Figure 3 This diagram illustrates the 60% and 90% confidence intervals of the standard normal distribution Z-index provided in the embodiments of this specification.

[0063] Figure 4 This is a block diagram of an electronic device provided in the embodiments of this specification. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0065] In related technologies, with the intensifying impact of global climate change and the increasing frequency of extreme hydrological events (such as regional droughts and floods), the stability of water resource supply in a single river basin faces severe challenges. Against this backdrop, constructing inter-basin water transfer projects to achieve mutual assistance and complementarity of water resources between different river basins has become a key approach to ensuring regional water supply security and improving the efficiency of comprehensive water resource utilization. The core decision-making basis for inter-basin water resource allocation is the accurate assessment of the spatiotemporal complementarity of runoff inflows between different river basins.

[0066] However, in related technologies, the assessment methods for cross-basin runoff complementarity have significant limitations, making it difficult for the assessment results to effectively support accurate scheduling decisions.

[0067] Specifically, firstly, the assessment dimensions are one-sided. Related techniques often focus on single-dimensional analysis. For example, they only focus on the temporal synchronicity of runoff processes (e.g., simple statistics on the frequency of wet and dry seasons) or only examine spatial differences in total volume (e.g., comparisons of average annual runoff), lacking a framework that systematically integrates the two inherently complementary dimensions of temporal synchronicity and spatial variability. This one-sided assessment cannot fully characterize the complex dynamic complementary relationships between watersheds and fails to reflect the overall robustness and regulatory potential of the water resources system.

[0068] Secondly, the evaluation process is highly subjective. Due to the lack of a complete and quantifiable indicator system, related technologies often rely on expert experience or arbitrarily set thresholds when determining the importance (weight) of indicators and assessing their complementarity (grading). This subjectivity leads to varying evaluation results from person to person and case to case, lacking an objective and unified standard. This results in incomparable evaluation conclusions between different watershed combinations, making it difficult to provide a reliable prioritization basis for water diversion project planning at the macro level.

[0069] Ultimately, these shortcomings collectively lead to the technical problem of weak decision support capabilities. One-sided assessment conclusions and subjective classification results cannot provide accurate and reliable data support for core scheduling decisions such as when to transfer water, where to transfer water, and the scale of water transfer, thus affecting the accuracy of cross-basin water resource allocation schemes.

[0070] like Figure 1 As shown in this embodiment, a cross-basin hierarchical scheduling method based on runoff spatiotemporal complementarity assessment is provided. The method can be executed by a locally deployed server or workstation, which can act as a computing node, directly receiving runoff time-series data from various watershed hydrological monitoring stations or retrieving historical runoff time-series data from a local database. Alternatively, the method can be executed by a cloud computing platform, where users can upload runoff time-series data via a client. The cloud computing platform can then perform analysis and return the final analysis results (complementarity level) to the user.

[0071] The method may include:

[0072] Step S10: Obtain runoff time series data for multiple watersheds within a preset scheduling period.

[0073] In this embodiment, the acquisition action can be represented as the execution subject (e.g., a workstation or server) receiving runoff time series data of multiple watersheds within a preset scheduling period by calling a local database, receiving from a remote hydrological monitoring server, or acquiring data in real time through a sensor network.

[0074] In this embodiment, the runoff time series data can be represented as a dataset continuously recorded at fixed time intervals (e.g., days and months) within the preset scheduling period, reflecting the flow magnitude at the outlet sections of each watershed. The preset scheduling period can be set according to actual analysis needs; for example, it can be a complete hydrological year, a dry season or a wet season, or a long-term series consisting of multiple consecutive years, etc. The multiple watersheds can be represented as two or more independent hydrological units that are hydraulically connected or have the potential for engineering connectivity.

[0075] Step S12: Based on the runoff time series data, calculate the time dimension index and the spatial dimension index; wherein, the time dimension index is used to assess the temporal synchronicity of runoff changes between watersheds, and the spatial dimension index is used to assess the differences in runoff volume surplus and deficit and fluctuation between watersheds.

[0076] In this embodiment, the time dimension index can be used to assess the consistency or inconsistency of runoff processes in different watersheds over time. Higher inconsistency indicates greater potential for complementarity in the time dimension. This time dimension index may include: a consistency index for abundance and scarcity, which can quantify the degree to which different watersheds are in the same or opposite states of abundance and scarcity during the same period. This time dimension index may also include: a consistency index for change trends, which can quantify the synchronicity of runoff volume increases or decreases in different watersheds during adjacent periods.

[0077] In this embodiment, the spatial dimension index can be used to assess the differences in water surplus and deficit and fluctuation characteristics between different watersheds due to geographical and climatic variations. The greater the difference, the greater the potential for spatial dimension complementarity. The spatial dimension index can include a surplus difference index and a fluctuation difference index. The surplus difference index can be used to quantify the potential compensating capacity of a watershed's water surplus for a watershed's water deficit in another watershed. The fluctuation difference index can be used to quantify the degree to which runoff fluctuations in different watersheds tend to stabilize after being superimposed.

[0078] Step S14: Based on the preset entropy weight method, determine the objective weight corresponding to each indicator according to the numerical distribution of the time dimension indicator and the spatial dimension indicator.

[0079] In this embodiment, the preset entropy weighting method can be expressed as a method of objectively assigning index weights based on the dispersion of the data itself. Specifically, the entropy weighting method can be interpreted as follows: the greater the difference in the value of an index among different samples (i.e., the higher the dispersion), the greater the amount of information that the index can provide in the comprehensive evaluation, and the higher its weight should be. More specifically, step S14 may include: firstly, standardizing the original values ​​of each index calculated in step S12 to eliminate the influence of dimensions. Then, calculating the probability distribution of all sample values ​​under each index, and thus obtaining the information entropy of the index. The smaller the information entropy, the greater the difference coefficient of the index, and finally determining its objective weight based on the difference coefficient of each index.

[0080] Step S16: Based on the objective weights, perform a weighted summation of the standardized index values ​​to generate a spatiotemporal comprehensive complementarity index.

[0081] In this embodiment, the weighted summation can be expressed as multiplying each standardized index value by its objective weight determined by the entropy weight method in step S14, and then summing all the product results to obtain a comprehensive value, namely the spatiotemporal comprehensive complementarity index. This spatiotemporal comprehensive complementarity index can be expressed as a quantitative value that simultaneously reflects the complementary characteristics of both time and space dimensions. The level of this index directly characterizes the comprehensive magnitude of the runoff complementarity potential within the analyzed multiple watersheds as a whole system.

[0082] Step S18: Convert the spatiotemporal comprehensive complementarity index into a Z-index that follows a standard normal distribution.

[0083] In this embodiment, the skewness coefficient of the spatiotemporal comprehensive complementarity index sequence can be calculated to quantify its distribution asymmetry. Then, based on this skewness coefficient, the original index value is converted into a Z-index value through a normalization transformation function. It is understood that since the directly calculated spatiotemporal comprehensive complementarity index may not follow a normal distribution, it is not convenient for unified hierarchical evaluation. Therefore, this embodiment transforms it into a Z-index that follows (or approximately follows) a standard normal distribution.

[0084] Step S110: Determine the runoff complementarity level among the multiple watersheds based on the quantile interval of the Z-index under the standard normal distribution.

[0085] In this embodiment, the quantile interval can be pre-defined based on the theoretical characteristics of the standard normal distribution. For example, it can be set that when the Z-index is greater than or equal to 1.645, it corresponds to a very strong complementarity level; when the Z-index is between 0.842 and 1.645, it corresponds to a strong complementarity level; when the Z-index is between -0.842 and 0.842, it corresponds to a neutral complementarity level; when the Z-index is between -1.645 and -0.842, it corresponds to a weak complementarity level; and when the Z-index is less than or equal to -1.645, it corresponds to a very weak complementarity level.

[0086] Step S112: Based on the runoff complementarity level, generate scheduling control signals for controlling inter-basin water transfer projects.

[0087] In this embodiment, step S112 may include:

[0088] Step S1121: When the runoff complementarity level is extremely strong complementarity, a first type of scheduling control signal is generated. The first type of scheduling control signal is used to indicate operation in an active scheduling mode. The water transfer direction is determined based on the runoff surplus difference analysis results, and the water transfer volume is determined based on the theoretical upper limit calculated by the runoff complementarity rate.

[0089] It is understandable that the extremely strong complementarity level can indicate that the runoff between multiple watersheds has extremely high complementarity potential in the temporal and spatial dimensions. That is, the runoff abundance and scarcity states between watersheds are highly asynchronous, the water surplus and shortage are significantly opposed and the fluctuation characteristics cancel each other out very well, thus providing optimal conditions for large-scale inter-watershed water transfer.

[0090] In this embodiment, the first type of scheduling control signal can be an electronic control command or a digital signal, which can directly drive or instruct inter-basin water transfer projects (e.g., pumps, gates, water pipelines, etc.) to operate in an active scheduling mode. The active scheduling mode can be a high-intensity, high-response scheduling strategy that maximizes the utilization of water complementarity potential between basins. Through rapid, large-scale water transfer operations, it achieves efficient redistribution of water resources to prioritize power generation and water supply needs while also considering flood control safety.

[0091] In one possible and specific implementation, the direction of water diversion can be determined based on the results of runoff surplus difference analysis. This runoff surplus difference analysis can be expressed as identifying the spatiotemporal distribution of water surplus and shortage among watersheds by calculating the runoff deviation value (i.e., the difference between the runoff volume in each time period and the multi-year average runoff volume) for each watershed. Specifically, the direction of water diversion can be determined as flowing from watersheds with positive runoff deviation values ​​(indicating water surplus) to watersheds with negative runoff deviation values ​​(indicating water shortage). For example, in the analysis, if the runoff deviation value of watershed A is consistently positive, while the runoff deviation value of watershed B is consistently negative, then the direction of water diversion is set to flow from watershed A to watershed B. Furthermore, based on the magnitude of the surplus difference index (runoff complementarity rate), the pair of watersheds with the largest surplus and the most significant shortage can be preferentially selected as the water diversion path. Geographic Information System (GIS) data can also be used to optimize the water conveyance route to reduce engineering costs.

[0092] Specifically, the water transfer volume can be determined based on the theoretical upper limit calculated from the runoff complementarity rate. The runoff complementarity rate is a quantitative indicator representing the proportion of compensable water volume between watersheds to the total runoff volume during the scheduling period. The theoretical upper limit can be expressed as the maximum value of the calculated runoff complementarity rate at that complementarity level, i.e., as close as possible to the total compensable water volume, to fully utilize the complementarity potential. More specifically, the water transfer volume can be calculated in the following ways:

[0093] First, the compensable water volume for all time periods within the scheduling period is accumulated (i.e., the smaller absolute value of the deviation values ​​when the runoff deviation values ​​of the two watersheds have opposite signs). Then, this accumulated value is used as the theoretical upper limit of the water diversion volume. The water diversion volume can also be fine-tuned according to the engineering capacity (pipeline water conveyance capacity, reservoir regulation capacity).

[0094] Specifically, dispatch instructions can be issued on a ten-day or monthly timescale. Once a significant surplus or deficit is detected (e.g., a sudden change in runoff detected by a sensor network), signal generation is immediately triggered. Extending the response mechanism, it may include an automatic control loop, where dispatch control signals are directly sent to actuators to achieve unattended operation.

[0095] Step S1122: When the runoff complementarity level is strong complementarity, a second type of scheduling control signal is generated. The second type of scheduling control signal is used to indicate the operation of the optimized scheduling mode. The water transfer direction is determined based on the consistency between abundant and scarce periods and the difference in surplus. The water transfer volume is determined based on a preset ratio of the potential calculated by the runoff complementarity rate.

[0096] In this embodiment, it is understood that a strong complementarity level indicates a high potential for runoff complementarity between watersheds, but it is not as significant as that of an extremely strong complementarity level, and may involve certain uncertainties or fluctuations. The second type of scheduling control signal can be an optimization control command, which can instruct water diversion projects to operate in an optimized scheduling mode. This mode can balance complementary utilization with system risks and achieve robust scheduling through comprehensive multi-indicator decision-making.

[0097] The optimized scheduling mode can fully utilize complementary potential while reserving buffer space for the system to cope with hydrological uncertainties and ensure the reliability and adaptability of scheduling decisions.

[0098] In a possible and specific implementation plan, the direction of water diversion can be determined based on the consistency of runoff abundance and scarcity and the difference in surplus. The consistency of runoff abundance and scarcity represents the degree of consistency between the runoff abundance and scarcity states of different watersheds; a smaller value indicates higher asynchrony between abundance and scarcity, which is more conducive to water diversion. The decision-making process for the direction of water diversion can be as follows: First, possible water diversion paths are initially determined through surplus difference analysis (e.g., from surplus watersheds to shortage watersheds). Then, a consistency index is introduced for correction. If the consistency is low, the water diversion direction is confirmed as feasible. If the consistency is high, the water diversion direction can be adjusted or the scale of water diversion can be reduced. For example, the consistency values ​​of multiple watershed pairs can be compared, and the watershed pair with the smallest value can be selected as the priority for water diversion. More specifically, multi-criteria decision-making methods can be used, such as fuzzy logic or weighted scoring, using consistency and surplus difference as input variables to output the optimal water diversion direction. Alternatively, historical scheduling data can be combined with machine learning models to predict the effects of different directions.

[0099] In one possible and specific implementation, the water diversion volume can be determined based on a preset proportion of the potential calculated using the runoff complementarity rate. Specifically, the potential calculated using the runoff complementarity rate can be expressed as the theoretical value of the compensable water volume obtained through runoff complementarity rate calculation. The preset proportion can be a configurable parameter, set between 60% and 80%, used to control the water diversion volume from exceeding a certain proportion of the potential to preserve buffer capacity. More specifically, the water diversion volume can be obtained by multiplying the theoretical compensable water volume calculated using the runoff complementarity rate by the preset proportion. The preset proportion can also be dynamically adjusted; for example, different proportions can be set according to seasonal changes (upper limit for rainy season, lower limit for dry season). The issuance of instructions for this optimized scheduling mode can also be on a monthly or quarterly scale, combined with dynamic fine-tuning.

[0100] Step S1123: When the runoff complementarity level is flat complementarity, a third type of scheduling control signal is generated. The third type of scheduling control signal is used to indicate operation in a conservative scheduling mode. The water transfer decision is based on fluctuation differences, and the water transfer volume is limited to below a preset threshold of complementarity potential.

[0101] In this embodiment, "balanced complementarity" can indicate that the potential for runoff complementarity between watersheds is generally low, meaning the complementarity characteristics are not significant. Therefore, the third type of dispatch control signal can be a conservative control command, which instructs water transfer projects to operate in a conservative dispatch mode, maintaining system stability through small-scale, intermittent water transfers. In other words, the conservative dispatch mode prioritizes ensuring basic water supply security, and water transfer operations are only initiated when there is a brief, localized opportunity for complementarity in the water situation.

[0102] In one possible and specific implementation, water diversion decisions can be made based on fluctuation differences. Fluctuation differences can represent the degree to which runoff fluctuations between watersheds cancel each other out; the smaller the value, the better the complementarity of the fluctuations. The water diversion decision-making process can be as follows: real-time monitoring of runoff changes in each watershed (runoff difference between adjacent time periods); when multiple watersheds are detected to have opposite changes (i.e., one increasing and one decreasing) in the same time period, it is determined that there is a complementary opportunity, and a water diversion signal is triggered. For example, if the runoff in watershed A increases while the runoff in watershed B decreases, a small-flow diversion signal is generated, directed from A to B.

[0103] More specifically, sliding window analysis can be used, which assesses the mean or variance of fluctuation differences over multiple consecutive time periods, and only initiates water adjustment when the value falls below a threshold. Alternatively, time series analysis can be combined to identify the duration of inverse correlations in fluctuations, thereby determining the timing of water adjustments.

[0104] The water diversion volume is limited to below a preset threshold of the complementary potential. The complementary potential can be calculated based on the runoff complementarity rate, and the preset threshold can be set to below 30% of the complementary potential to ensure that the water diversion volume is minimal and avoids disrupting the watershed's own supply and demand balance. Specifically, the water diversion volume can be obtained by multiplying the theoretical value calculated from the complementary potential by a conservative coefficient (e.g., 0.2).

[0105] The conservative scheduling mode's instructions can be based on quarterly or annual planning, with priority given to ensuring water supply security. That is, its scheduling signals are generated at a lower frequency, focusing on long-term planning rather than real-time response.

[0106] Step S1124: When the runoff complementarity level is weak complementarity or very weak complementarity, a fourth type of scheduling control signal is generated. The fourth type of scheduling control signal is used to indicate operation in defensive scheduling mode; wherein, large-scale inter-basin water transfer is suspended, and small-flow adjustment is only initiated in emergency situations.

[0107] In this embodiment, weak complementarity or very weak complementarity can indicate that the runoff complementarity potential between watersheds is extremely low, meaning that the runoff processes are highly synchronized or have only minor differences, and forced water transfer may disrupt the water balance of each watershed. Therefore, the fourth type of scheduling control signal can be a defensive control command, used to instruct water transfer projects to operate in a defensive scheduling mode. This mode can avoid inter-basin water transfer and prioritize relying on local resources to meet demand. The defensive scheduling mode can shift the focus of water resource management from interconnection to independent scheduling to minimize system risk. It is understood that suspending large-scale inter-basin water transfer means that regular water transfer commands are suppressed, and water transfer projects are in a standby or low-speed operation state. The large-scale water transfer can refer to actions where the water transfer volume exceeds a certain proportion (e.g., 10%) of the complementarity potential. The generation of such signals can be prohibited by interlocking logic unless emergency conditions are met.

[0108] The phrase "activating small-flow water diversion only in emergency situations" refers to generating a very small-scale water diversion signal when an extreme event (such as a water supply crisis in a river basin) occurs. The emergency situation can be determined using a threshold, for example, when the water level in the river basin reservoir is below the dead water level or when the water demand continuously exceeds the local water supply capacity for a certain number of days. The diversion volume of the small-flow water diversion is extremely low, and can not exceed 10% of the complementary potential. Specifically, the diversion volume can be set to a fixed minimum value (e.g., several cubic meters per second) or dynamically calculated based on the severity of the crisis.

[0109] The defense scheduling mode can be independent scheduling within the basin, that is, giving priority to the use of local backup water sources (e.g., groundwater, reclaimed water) and emergency measures (water conservation orders, temporary water storage).

[0110] This embodiment systematically quantifies complementary characteristics from two physical dimensions, time and space, overcoming the one-sidedness of single-index evaluation. Its comprehensive index can more realistically reflect the synergistic effect of complex watershed systems, making the analysis results of different watershed groups and different time periods consistent and reliable comparison benchmarks.

[0111] This embodiment automatically determines the index weights using the entropy weight method and achieves standardized grading through Z-index transformation, eliminating the subjectivity and arbitrariness of manually setting weights and thresholds, and enabling the analysis process to be executed automatically and in a standardized manner by computing devices.

[0112] The spatiotemporal comprehensive complementarity index and complementarity level output in this embodiment can be used as precise and quantitative input parameters to directly serve subsequent automated decision-making systems such as cross-basin water resource scheduling models and hydropower system optimization operation programs, thereby improving the collaborative operation efficiency and decision-making intelligence level of the entire water resource management system.

[0113] In summary, this embodiment constructs a quantitative index system with time and space dimensions, uses the entropy weight method to objectively assign weights and synthesize a comprehensive index, and then performs normalization transformation and standard classification. This achieves a technological leap from subjective experience-based judgment to objective machine evaluation of cross-basin runoff complementarity characteristics, providing standard and reliable technical parameter inputs for core water resource scheduling.

[0114] In some implementations, the time dimension index includes a first time dimension index and a second time dimension index; wherein, the first time dimension index is a consistency index for assessing the consistency of runoff abundance and scarcity between watersheds, and the second time dimension index is a trend consistency index for assessing the consistency of runoff change trends between watersheds.

[0115] In this embodiment, the first time dimension index can quantify the consistency of runoff abundance / dryness states in different watersheds within the same time period. Specifically, firstly, the criteria for classifying abundance / dryness levels for each watershed can be determined based on runoff time series data. More specifically, the cumulative runoff frequency method can be used to classify runoff sequences into five levels: extremely low, slightly low, normal, slightly high, and extremely high. Alternatively, the mean-standard deviation method can be used, with the multi-year average runoff as a benchmark, to classify levels according to the degree of deviation. Then, a consistency measurement model for abundance / dryness states can be established. Specifically, the degree of consistency can be characterized by statistically analyzing the proportion of time periods in which two watersheds are at the same abundance / dryness level within the same time period out of the total number of time periods. A more refined correlation algorithm can also be used, which considers not only completely consistent time periods but also time periods with adjacent or opposing levels, and obtains a comprehensive consistency index through weighted calculation.

[0116] In this embodiment, the trend consistency index can be used to assess the synchronicity of runoff trends in different watersheds. Specifically, the runoff change rate between adjacent time periods can be calculated first. For example, a relative change rate method can be used, which is the difference between the runoff volume in the later time period and the runoff volume in the earlier time period divided by the runoff volume in the earlier time period. Alternatively, an absolute change rate method can be used to calculate the runoff change rate between adjacent time periods; for example, the difference in runoff volume can be used directly. Then, the trend consistency index can be obtained by comparing the direction and magnitude of the runoff change rates between adjacent time periods. For example, the average of the absolute values ​​of the differences or further analysis of the contribution ratios of unidirectional and inverse changes can comprehensively characterize the consistency level of the two runoff sequences in terms of trend.

[0117] The spatial dimension indicators include a first spatial dimension indicator and a second spatial dimension indicator; wherein, the first spatial dimension indicator is a surplus difference indicator used to assess the runoff surplus compensation capacity between watersheds, and the second spatial dimension indicator is a fluctuation difference indicator used to assess the degree of runoff fluctuation offsetting between watersheds.

[0118] In this embodiment, the first spatial dimension index can be represented as a parameter used to quantify the potential compensatory capacity of a watershed's water surplus for a watershed's water deficit. It can assess the matching characteristics of water surplus and deficit between spatially separated watersheds. Specifically, firstly, the deviation of runoff from this benchmark can be calculated for each time period, using the multi-year average runoff of each watershed as a baseline. Then, periods where one watershed has a positive deviation (surplus) and another has a negative deviation (shortage) are identified. Within these periods, the smaller of the absolute values ​​of the two deviations is taken as the theoretically compensable water volume for that period. Finally, the compensable water volumes for all periods within the scheduling period are summed and compared with the sum of the total runoff of all watersheds during the same period. The resulting ratio is the index value, which intuitively reflects the potential scale of surplus complementarity.

[0119] In this embodiment, the second spatial dimension index, namely the fluctuation difference index, can be represented as a parameter used to quantify the degree to which the runoff fluctuation processes of multiple watersheds cancel each other out, thereby stabilizing the total system input. Specifically, firstly, the runoff change of each watershed in each time period during the scheduling period can be calculated. Then, for each time period, the algebraic sum of the runoff changes of all watersheds in that time period is calculated. Finally, the average value of the algebraic sum and its absolute value over the entire scheduling period is calculated. The smaller this average value, the better the spatial complementarity of the runoff fluctuations of each watershed, and the more stable the total system inflow.

[0120] In some implementations, the step of calculating time dimension indicators and spatial dimension indicators based on the runoff time series data includes: the step of calculating a first time dimension indicator and the step of calculating a second time dimension indicator;

[0121] Step S122: The step of calculating the first time dimension index includes:

[0122] Step S1221: Based on the preset runoff cumulative frequency threshold, the runoff data for each time period are divided into multiple wet and dry levels.

[0123] In this embodiment, the preset runoff cumulative frequency threshold can be expressed as a quantile value set according to long-term hydrological statistical characteristics.

[0124] In this embodiment, the probability distribution of runoff can first be determined. One specific implementation method can employ kernel density estimation. This method obtains a smooth runoff probability density curve by nonparametric fitting of historical runoff data. This curve avoids the subjectivity of traditional histogram grouping and can more accurately reflect the overall distribution pattern of the quantity, such as... Figure 2As shown in the figure above, the smooth black curve is the probability density function curve obtained from kernel density estimation. Then, the cumulative frequency threshold is determined based on the probability distribution. Based on the obtained probability density function, its cumulative distribution function (CDF) can be calculated through integration, i.e. Figure 2 The figure below shows an S-shaped curve. The preset runoff cumulative frequency thresholds (e.g., 12.5%, 37.5%, 62.5%, 87.5%) are key quantiles selected on this cumulative frequency curve. Finally, the thresholds are applied for grading. After determining the flow thresholds corresponding to each grade, the specific runoff volume for any given time period can be compared with these thresholds to classify it into the corresponding abundant / dry grade. For example, if the runoff volume for a certain time period falls within the interval formed by the flow values ​​of the 12.5% ​​and 37.5% quantiles, then that time period is classified as a dry grade.

[0125] In one possible and specific implementation plan, the quintile method can be used for classification. The runoff sequence can be divided into five levels using the four cumulative frequencies of 12.5%, 37.5%, 62.5%, and 87.5% as the dividing points: extremely low (0-12.5%), moderately low (12.5%-37.5%), normal (37.5%-62.5%), moderately high (62.5%-87.5%), and extremely high (87.5%-100%).

[0126] In one possible and specific implementation plan, a ternary method can also be used for simplified division, with 33.3% and 66.7% as the dividing points, dividing the runoff into three basic levels: low water, normal water, and high water.

[0127] It is understandable that the specific value of the cumulative frequency threshold can be adaptively adjusted and set in advance according to the hydrological characteristics and analysis accuracy requirements of different regions.

[0128] Step S1222: Based on the abundance and scarcity levels of the runoff sequences of the two watersheds at each time period, the abundance and scarcity consistency index is calculated according to the number of time periods with the same level, the number of time periods with a level difference of the first preset level, the number of time periods with a level difference of the second preset level, and the number of time periods with opposing levels.

[0129] In this embodiment, grade difference statistics can be performed. For the abundance / dampness grade sequences of the two watersheds, the grade differences are compared time periods. That is, the number of time periods with the same grade, the number of time periods with a grade difference of the first preset grade, the number of time periods with a grade difference of the second preset grade, and the number of time periods with opposing grades are counted.

[0130] In this embodiment, the first preset level can be represented as a level difference of one level, the second preset level can be represented as a level difference of two levels, and the opposing level can be represented as a level difference of three levels or more.

[0131] In a possible and specific implementation plan, firstly, all time periods within the entire scheduling period can be traversed to statistically compare the combinations of abundance and scarcity levels of the two watersheds, and the number of the following four types of time periods can be counted:

[0132] The number of time periods with the same classification can be expressed as the total number of time periods in which two watersheds are classified into the exact same classification during the same time period.

[0133] The number of time periods with a grade difference of the first preset level can be expressed as the total number of time periods in which the grades of two watersheds differ by one level. For example, one watershed is relatively abundant while the other is at normal water level, or one is relatively dry while the other is at normal water level.

[0134] The number of time periods with a grade difference of the second preset level can be expressed as the total number of time periods in which the grades of two watersheds differ by two levels. For example, one watershed is exceptionally wet while the other is at normal water levels, or one is exceptionally dry while the other is moderately wet.

[0135] The number of time periods with opposing grades can be expressed as the total number of time periods in which the grades of two watersheds are in completely opposite states. It can be defined as a combination of grades with differences of three or more, such as when exceptionally abundant water meets exceptionally scarce water, or when moderately abundant water meets moderately scarce water.

[0136] Then, based on the above statistics, the final consistency index of abundance and scarcity can be obtained by calculating the correlation degree. Specifically, the proportion of time periods with the same level to the total number of time periods can be defined as the degree of uniformity. The proportion of time periods with a level difference of the first preset level to the total number of time periods can be multiplied by a preset first identification coefficient between 0 and 1 (e.g., 0.5) as the first degree of difference. The proportion of time periods with a level difference of the second preset level to the total number of time periods can be multiplied by a preset second identification coefficient less than the first identification coefficient (e.g., 0.25) as the second degree of difference. The proportion of time periods with opposite levels to the total number of time periods can be multiplied by a preset negative opposite identification coefficient (e.g., -1) as the opposite degree of difference.

[0137] Finally, by adding the aforementioned similarity, first difference component, second difference component, and opposition component, the result is the abundance-dryness consistency index. The closer this index value is to 1, the higher the synchronicity of abundance and drought between the two watersheds; the closer it is to -1, the higher the asynchronicity of abundance and drought, and the greater the potential for complementarity.

[0138] Step S124: The step of calculating the second time dimension index includes:

[0139] Step S1241: Calculate the rate of change of the runoff time series data in adjacent time periods.

[0140] In this embodiment, the specific length of the adjacent time periods depends on the scale of the original runoff time series. If the series is daily runoff data, then adjacent time periods can be two consecutive days; if it is monthly runoff data, then adjacent time periods can be two consecutive months.

[0141] In this embodiment, all consecutive time periods within the entire scheduling period can be traversed to form a rate of change sequence corresponding to the original runoff sequence length minus one.

[0142] In one possible and specific implementation, a relative rate of change method can be used to calculate the rate of change of the runoff time series data in adjacent time periods. This can be achieved by subtracting the runoff of the previous time period from the runoff of the later time period, and then dividing by the runoff of the previous time period; the result is expressed as a percentage.

[0143] In one possible and specific implementation, the absolute change method can be used to calculate the rate of change of the runoff time series data in adjacent time periods. The difference in runoff volume between adjacent time periods can be calculated directly.

[0144] Step S1242: Calculate the consistency index of change trend based on the difference in the runoff change rate of the two watersheds in the corresponding time period.

[0145] In this embodiment, the consistency of the changing trends of the two watersheds can be quantified based on the rate of change sequences of the two watersheds calculated in step S1241. The difference is used to measure the similarity of the runoff change behavior of the two watersheds within the same time period.

[0146] In one possible and specific implementation, the trend consistency index can be calculated as follows:

[0147] First, calculate the absolute value of the difference in the rate of change between the two watersheds at each corresponding time period, i.e., |A{1,t}-A{2,t}|, where A{1,t} and A{2,t} represent the rate of change of watershed 1 and watershed 2 at the t-th time period, respectively.

[0148] Then, the arithmetic mean of the calculated absolute values ​​of the differences is obtained.

[0149] Finally, the consistency index of the trend of change can be defined based on the average value. Since a smaller average value indicates higher consistency, the average value can be directly used as the index, in which case it is a negative index. In another implementation, it can also be converted into a positive index, for example, by mapping it to a range of 0 to 1 through linear or nonlinear transformation, where 1 represents a completely consistent trend of change.

[0150] In one possible and specific implementation plan, the differences in the direction of change can be further distinguished. Specifically, the entire time period can be divided into two subsets: a subset of changes in the same direction (the change rates of the two watersheds have the same sign) and a subset of changes in opposite directions (the change rates of the two watersheds have opposite signs). The average of the absolute values ​​of the differences in the change rates within each of the two subsets is calculated, and the weight of each subset's time period relative to the total number of time periods is considered. Finally, a comprehensive consistency index is synthesized by weighted averaging.

[0151] In some implementations, the step of calculating time dimension indicators and spatial dimension indicators based on the runoff time series data includes: the step of calculating a first spatial dimension indicator and the step of calculating a second spatial dimension indicator.

[0152] Step S126: The step of calculating the first spatial dimension index includes:

[0153] Step S1261: Based on the preset multi-year average runoff of each watershed, calculate its runoff deviation value in each time period.

[0154] In this embodiment, this step determines the degree of deviation of runoff volume in each time period from the long-term average level, thereby identifying a water surplus or shortage. The preset multi-year average runoff volume can be a long-term average calculated based on historical hydrological data.

[0155] Specifically, the runoff time series data for each watershed within the preset scheduling period can be obtained first, and then the runoff deviation value can be calculated for each time period. More specifically, the runoff deviation value can be calculated using the absolute deviation method, which involves subtracting the multi-year average runoff of the watershed from the actual runoff for each time period. A positive value indicates a water surplus, while a negative value indicates a water shortage. Alternatively, the relative deviation method can be used, which divides the absolute deviation value by the multi-year average runoff and expresses the degree of deviation as a percentage.

[0156] Step S1262: For each time period, if the runoff deviation values ​​of the two watersheds have opposite signs, the compensable water volume for that time period is determined based on the absolute value of the runoff deviation values ​​of the two watersheds.

[0157] In this embodiment, this step can identify opportunities for complementary water surpluses and deficits between watersheds and quantify the potential compensable water volume. Specifically, the signs of the runoff deviation values ​​of the two watersheds can be checked on a time-by-time basis: when the signs are opposite (i.e., one watershed is in surplus and the other is in deficit), complementary potential is considered to exist.

[0158] In this embodiment, the compensable water volume can be determined based on the minimum value principle. That is, the smaller of the absolute values ​​of the runoff deviations of the two watersheds can be taken as the actual dispatchable water volume for that period. For example, if the deviation of watershed A is +100 cubic meters per second (surplus) and the deviation of watershed B is -80 cubic meters per second (shortage), then the compensable water volume is 80 cubic meters per second. This principle can ensure the feasibility of the compensable water volume and avoid over-dispatch.

[0159] In some cases, if the absolute values ​​of the deviations differ significantly, a threshold limit can be introduced, so that the deviation is only included in the calculation when it exceeds a preset critical value (e.g., 10% of the multi-year average runoff), in order to exclude the influence of small fluctuations.

[0160] For time periods with the same sign or any deviation value of zero, the compensable water volume can be directly recorded as 0, indicating that there is no complementary potential in that time period.

[0161] Step S1263: Calculate the surplus difference index based on the ratio of the sum of compensable water volume for all time periods during the scheduling period to the sum of the total runoff of the two watersheds.

[0162] In this embodiment, this step aggregates the compensable water volume at the time-period level into an overall complementarity index within the scheduling period. Specifically, the compensable water volume for all time periods is first summed to obtain the total compensable water volume. Then, the total runoff of the two watersheds during each time period within the scheduling period can be calculated. The ratio is calculated by dividing the total compensable water volume by the total runoff, and can be expressed as a percentage.

[0163] In this embodiment, the larger the value of the surplus difference index, the higher the water volume complementarity potential between watersheds. For example, if the total compensable water volume is 5,000 cubic meters and the total runoff is 100,000 cubic meters, the index value is 5%. In this embodiment, the ratio can also be normalized to fall within the range of 0 to 1, facilitating integration with other indicators. In this embodiment, to eliminate the influence of extreme values, a sliding window method or a weighted average method can be used to process the time-series data to ensure the stability of the index.

[0164] Step S128: The step of calculating the second spatial dimension index includes:

[0165] Step S1281: Calculate the runoff changes of the multiple watersheds during each time period within the scheduling period.

[0166] In this embodiment, this step quantifies the fluctuations in runoff in each watershed between adjacent time periods. Specifically, for short-term fluctuation analysis, daily or decadal time periods can be used as the adjacent time periods. For medium- to long-term trend analysis, monthly or yearly time periods can be used.

[0167] In this embodiment, the runoff changes in the multiple watersheds during the scheduling period can be calculated based on the absolute change calculation method. The change can be obtained by directly subtracting the runoff of the previous period from the runoff of the later period. Positive values ​​indicate an increase in runoff, and negative values ​​indicate a decrease in runoff.

[0168] In this embodiment, the runoff changes in the multiple watersheds during each time period within the scheduling period can be calculated based on the relative rate of change method. The absolute change can be divided by the runoff volume of the previous time period to obtain the rate of change in percentage form.

[0169] Step S1282: Calculate the fluctuation difference index based on the algebraic sum of the runoff changes of the multiple watersheds in the same period.

[0170] In this embodiment, this step can quantify the degree of complementarity by the superposition effect of fluctuation processes between watersheds. Specifically, the algebraic sum of the changes in runoff of all watersheds in each time period can be calculated first. This algebraic sum is essentially the algebraic addition of the changes in runoff from each watershed in the same time period (considering the sign), yielding the net fluctuation of the system for that time period. Then, based on the algebraic sum, the fluctuation difference index is calculated. More specifically, the average of the absolute values ​​of the net fluctuation of the system in all time periods can be calculated. The smaller this value, the better the mutual cancellation effect of the fluctuations in each watershed, and the stronger the complementarity. The variance of the net fluctuation of the system can also be calculated; the smaller the variance, the more stable the system as a whole.

[0171] In some implementations, the step of determining the objective weights corresponding to each indicator based on the numerical distribution of the time dimension indicator and the spatial dimension indicator using a preset entropy weight method includes:

[0172] Step S142: Standardize the abundance / shortness consistency index, the trend consistency index, the surplus difference index, and the volatility difference index; wherein, the surplus difference index is standardized using a positive standardization method, and the abundance / shortness consistency index and the volatility difference index are standardized using a negative standardization method.

[0173] In this embodiment, the surplus difference index is positively correlated with the complementarity strength; that is, a larger index value indicates better complementarity. Therefore, in this embodiment, a positive standardization method can be used. Specifically, for all samples to be analyzed (e.g., different years or different river basin combinations), the maximum and minimum values ​​of the index can be determined from the original values. Then, for each original value of the index, the minimum value is subtracted, and then divided by the difference between the maximum and minimum values, thereby mapping the original value to the interval between 0 and 1.

[0174] In this embodiment, the values ​​of the abundance / abundance consistency index and the volatility difference index are negatively correlated with the complementarity strength; that is, the smaller the index value, the better the complementarity. Therefore, a negative standardization method can be used. Specifically, the maximum and minimum values ​​of all samples for this index can be determined first. Then, for each original value, the maximum value is subtracted from the original value, and then divided by the difference between the maximum and minimum values. This maps the processed value to the range of 0 to 1, where a larger value indicates better complementarity.

[0175] In this embodiment, when the trend consistency index is defined as a negative index (for example, the smaller the value, the more consistent the trend), the same negative standardization method as the abundance-shortage consistency index can be used; when it is defined as a positive index, the same positive standardization method as the surplus difference index can be used.

[0176] Step S144: Calculate the information entropy of each indicator based on the standardized indicator values.

[0177] In this implementation, the uncertainty or dispersion of the information provided by each indicator in the sample set can be quantified. The smaller the information entropy value, the stronger the discriminative power of the indicator, and the greater its weight should be given in the evaluation.

[0178] In this implementation, firstly, the probability distribution of the standardized values ​​of each sample under each indicator can be calculated. For example, the weight of each sample value can be calculated. Specifically, the standardized value of a single sample can be divided by the sum of the standardized values ​​of all samples for that indicator to obtain the probability of that sample value occurring. Then, the information entropy can be calculated based on the probability distribution. It can be understood that the calculation of information entropy is a weighted summation of the logarithms of the probabilities. First, the product of the natural logarithm of each probability value and the probability value can be calculated, then the product of all samples can be summed, and then the negative value can be taken. Finally, normalization is performed by dividing by the natural logarithm of the number of samples so that the entropy value falls within the range of [0, 1].

[0179] Step S146: Calculate the corresponding objective weight based on the information entropy of each indicator; where the greater the dispersion of the indicator, the smaller its information entropy, and the greater the weight assigned.

[0180] In this implementation, information entropy can be converted into usable weight coefficients, achieving objective weighting based on the degree of data dispersion. The principle of weight allocation is that the greater the dispersion of an indicator (the smaller the information entropy), the greater the weight assigned, because it contains more effective information. Specifically, firstly, the difference coefficient of each indicator can be calculated. The difference coefficient can be defined as 1 minus the information entropy value of the indicator. The larger the difference coefficient, the higher the degree of variation of the indicator among samples, and the stronger its discriminative ability. Then, weight normalization can be performed. The difference coefficient of each indicator can be divided by the sum of the difference coefficients of all indicators to obtain the final weight of each indicator.

[0181] In some embodiments, the step of converting the spatiotemporal comprehensive complementarity index into a Z-index that follows a standard normal distribution includes:

[0182] Step S182: Calculate the skewness coefficient of the spatiotemporal integrated complementarity index to quantify the asymmetry of its distribution.

[0183] In this embodiment, the skewness coefficient can be represented as a statistic characterizing the asymmetry of the probability distribution, and can be used to quantify the degree to which the data distribution deviates from symmetry. A positive skewness coefficient indicates that the distribution is skewed to the right, a negative skewness coefficient indicates that the distribution is skewed to the left, and a zero value indicates a symmetrical distribution.

[0184] In one possible and specific implementation, step S182 may include:

[0185] Step S1821: Add up all the sample values ​​of the spatiotemporal comprehensive complementarity index sequence, and then divide by the total number of samples to obtain the mean.

[0186] Step S1822: Calculate the square of the difference between each sample value and the mean, sum the squares, divide by the number of samples minus one, and then take the square root to obtain the standard deviation.

[0187] Step S1823: Calculate the cube of the difference between each sample value and the mean, then sum the cube values ​​and divide by the number of samples minus one to obtain the third central moment.

[0188] Step S1824: Divide the third central moment by the cube of the standard deviation to obtain the skewness coefficient value.

[0189] Step S184: Based on the skewness coefficient, perform a normalization transformation on the spatiotemporal comprehensive complementarity index to generate a Z-index that follows or approximately follows a standard normal distribution; wherein, the Z-index is used to eliminate the influence of the spatiotemporal comprehensive complementarity index on the complementarity level determination caused by the difference in the original distribution form.

[0190] In this embodiment, normalization transformation is a process of converting non-normally distributed data into data that follows or approximately follows a normal distribution using mathematical functions. This transformation is used to eliminate the skewness of the original data distribution, making it conform to the statistical characteristics of a standard normal distribution.

[0191] In this implementation, the original spatiotemporal comprehensive complementarity index sequence is first converted into a standardized variable, i.e., the sequence mean is subtracted from each value and then divided by the standard deviation. Then, the calculated skewness coefficients and the standardized variables are substituted into a normalization transformation function. This normalization transformation function can include mathematical operations such as logarithms, exponential functions, or power functions, and can adaptively adjust according to the actual skewness of the distribution. The new sequence output after this transformation is the Z-index.

[0192] In one possible and specific implementation, step S184 may include:

[0193] Step S1841: Determine the transform function type and transform strength parameters. Specifically, the direction of the transform function can be determined based on the sign of the calculated skewness coefficient values. When the skewness coefficient is positive, a transform function that compresses the right-tailed data is selected; when the skewness coefficient is negative, a transform function that compresses the left-tailed data is selected. The strength of the transform can be determined based on the absolute value of the skewness coefficient. The larger the absolute value of the skewness coefficient, the more severe the distribution skewness, requiring a stronger transform strength; the smaller the absolute value of the skewness coefficient, the weaker the transform strength.

[0194] Step S1842: Perform the function transformation calculation. Specifically, each spatiotemporal comprehensive complementarity index value can be substituted into a determined transformation function for calculation, resulting in a transformed intermediate numerical sequence.

[0195] Step S1843: Perform standardization. This involves calculating the mean and standard deviation of the transformed numerical sequence, subtracting the mean from each transformed value, and then dividing by the standard deviation to obtain the final Z-index sequence.

[0196] In some embodiments, the step of determining the runoff complementarity level among the plurality of watersheds based on the quantile interval of the Z-index under the standard normal distribution includes:

[0197] Step S1102: Compare the Z-index with a plurality of preset thresholds; wherein the plurality of thresholds include a first preset threshold, a second preset threshold, a third preset threshold and a fourth preset threshold whose values ​​decrease sequentially, and the thresholds are determined based on the quantiles of a standard normal distribution.

[0198] In this embodiment, the relative position of the Z-index within the standard normal distribution can be identified through numerical comparison operations, thereby providing a basis for grading. A quantile interval can be represented as the range of Z-values ​​corresponding to a specific probability under the standard normal distribution; for example, a 90% confidence interval corresponds to a Z-value of ±1.645. The quantile interval is used to define the rarity of an event, and in this embodiment, it serves as the statistical basis for grading.

[0199] In this embodiment, the preset threshold can be a fixed value pre-set based on quantiles, and may include: a first preset threshold, a second preset threshold, a third preset threshold, and a fourth preset threshold. The threshold values ​​decrease sequentially, corresponding to different probability boundaries. For example, the first preset threshold may correspond to the 95th quantile of the standard normal distribution (Z=1.645), the second preset threshold may correspond to the 80th quantile (Z=0.842), the third preset threshold may correspond to the 20th quantile (Z=-0.842), and the fourth preset threshold may correspond to the 5th quantile (Z=-1.645).

[0200] Step S1104: Based on the comparison results, the runoff complementarity levels are divided into five levels: extremely strong complementarity, strong complementarity, neutral complementarity, weak complementarity, and extremely weak complementarity. Specifically, when the Z-index value is greater than or equal to the first preset threshold, the complementarity level is determined to be extremely strong complementarity; when the Z-index value is less than the first preset threshold but greater than or equal to the second preset threshold, the complementarity level is determined to be strong complementarity; when the Z-index value is less than the second preset threshold but greater than or equal to the third preset threshold, the complementarity level is determined to be neutral complementarity; when the Z-index value is less than the third preset threshold but greater than or equal to the fourth preset threshold, the complementarity level is determined to be weak complementarity; and when the Z-index value is less than the fourth preset threshold, the complementarity level is determined to be extremely weak complementarity.

[0201] In this embodiment, the comparison results can be mapped to complementarity levels with clear hydrological significance, realizing the transformation from statistics to decision labels. The complementarity levels can include five categories: extremely strong complementarity, strong complementarity, neutral complementarity, weak complementarity, and extremely weak complementarity. These levels characterize the strength of the runoff complementarity potential between watersheds, with extremely strong complementarity representing the optimal complementarity and extremely weak complementarity representing the worst. The definition of the level is directly related to the quantile interval of the Z-index. For example, extremely strong complementarity corresponds to the Z-index falling in the extreme value region of the right tail of the distribution (e.g., Z ≥ 1.645), indicating that the complementarity of this watershed combination is significantly better than in most cases.

[0202] In one specific implementation plan, a cross-basin hierarchical scheduling method based on the assessment of runoff spatiotemporal complementarity characteristics is provided.

[0203] To scientifically assess the complementary potential and synergistic effects of runoff between different watersheds, this invention constructs a precise evaluation index system. This system systematically analyzes the complementary characteristics of cross-watershed runoff from two dimensions: the temporal consistency and spatial differences in runoff processes.

[0204] The time dimension analysis aims to determine the degree of synchronicity of runoff changes in different watersheds over time, i.e., whether they are "following each other closely" or "one rises while the other falls". The lower the synchronicity, the greater the potential for complementarity between watersheds over time. To this end, this invention evaluates the complementarity of the time dimension through runoff process consistency analysis, specifically including: (1) evaluation of consistency between wet and dry seasons: used to quantify the synergy of runoff changes in different watersheds during wet and dry seasons, characterized by the connectivity index; (2) evaluation of consistency of change trends: used to measure the similarity or difference in the long-term change trends of runoff in different watersheds, measured by the runoff consistency index.

[0205] Spatial dimension analysis aims to quantify the differences in water surplus and deficit and fluctuation characteristics of different watersheds caused by differences in geographical and climatic conditions. This is the material basis for realizing complementary scheduling in the form of "peak shaving and valley filling". This invention evaluates the complementary characteristics of spatial dimension through runoff difference analysis, specifically including: (1) Surplus difference evaluation: reflecting the ability of a watershed's water surplus to compensate for the water shortage of another watershed, which is quantified by the runoff complementarity rate; (2) Fluctuation difference evaluation: focusing on the ebb and flow relationship of runoff fluctuation processes in different watersheds, which is measured by the complementarity coefficient.

[0206] 1.1 Consistency Analysis of Runoff Processes

[0207] 1.1.1 Consistency Index between Abundance and Shortage

[0208] Consistency of runoff abundance and scarcity using correlation index The assessment is based on a comprehensive calculation of the consistency index of runoff abundance and scarcity between watersheds during the scheduling period. It is used to quantify the correlation between runoff abundance and scarcity between watersheds. The calculation formula is as follows:

[0209] ;

[0210] In the formula, It is the length in the runoff sequence. This represents the number of elements with the same sign in two sequences, therefore Indicates the same degree; This indicates the number of elements whose runoff abundance / shortness level difference is 1. The identifier representing difference 1, This indicates the proportion of difference 1 in the total population and its related characteristics; This indicates the number of elements whose runoff abundance / shortness level differs by 2. For identifiers with a difference level of 2, This indicates the proportion and related characteristics of elements with a runoff abundance / dampness level of 2 in the total population; Indicates the difference in runoff abundance / scarcity levels The number of elements at level 3 (opposite). These are identifiers that represent opposition. This indicates the proportion and related characteristics of opposing elements. , , .

[0211] The larger the value, the higher the consistency between watersheds in terms of abundance and scarcity, which is not conducive to runoff complementarity. The smaller the value, the more inconsistent the watershed conditions are, which is beneficial for runoff complementarity.

[0212] To further differentiate between high and low water inflow conditions, runoff accumulation frequencies of 12.5%, 37.5%, 62.5%, and 87.5% were used. Using this as the dividing point, runoff is divided into 5 levels according to its abundance or scarcity, as shown in Table 1.

[0213] Table 1. Classification of Abundance and Shortage Based on Cumulative Frequency

[0214]

[0215] 1.1.2 Consistency Index of Runoff Variation

[0216] This implementation plan proposes to use the runoff consistency index. To assess the consistency of runoff changes. For the runoff time series of the two watershed stations to be analyzed. and , The set is defined as follows:

[0217] ;

[0218] in, Represents flow sequence rate of change, Represents flow sequence The rate of change; Indicates the first of two flow sequences The absolute value of the difference between the changes; when At this time, the changes in the two sequences can be considered to be consistent; This indicates that the changes in the two sequences are inconsistent. This reflects the degree of inconsistency between the two time series changes. Specifically, Let be an element in set A, where the index i represents the i-th adjacent time period pair in the time series, that is, the time period consisting of the i-th time point to the (i+1)-th time point. The value of i ranges from n-1, because for a sequence including n time points, there are n-1 adjacent time periods. The subscript 1 in the text refers to the runoff time series of the first watershed. The subscript 'i' refers to the i-th adjacent time period. This represents the rate of change of the runoff time series of the first watershed within the i-th adjacent time period. The subscript 2 refers to the runoff time series of the second watershed. The subscript 'i' also refers to the i-th adjacent time period. It can represent the rate of change of the runoff time series of the second watershed in the i-th adjacent time period.

[0219] Therefore, to further analyze the degree of inconsistency in runoff changes between the two sequences, the following index is defined:

[0220] ;

[0221] in, A consistency index that measures the strength of the consistency between changes in two time series; Indicates the length of the time series; The closer the value is to 0, the higher the consistency of runoff changes between the two time series; conversely, the lower the consistency of runoff changes between the two series.

[0222] The two runoff time series are divided into two subsets: those with co-current variation and those with inverse variation.

[0223] ;

[0224] In the formula, This represents the absolute value of the difference in the rate of change of runoff between the two watersheds during the i-th time period. n is the total length of the runoff time series, i.e., the number of time points. The first subset, i.e., the subset that changes in the same direction. This is the second subset, namely the inverse variation subset.

[0225] By defining unidirectional consistency and antidirectional consistency indicators, and quantifying the degree of inconsistency caused by differences in unidirectional and antidirectional changes, it can be expressed as:

[0226] ;

[0227] In the formula, and Subsets and The sample size, and their proportion to the total sample size, are denoted as . and . and They represent subsets respectively and Consistency index of calculation.

[0228] 2.2 Analysis of runoff differences

[0229] In order to accurately quantify the complementary characteristics between watersheds, this invention constructs an evaluation index system from the perspectives of surplus difference and fluctuation difference to analyze the spatial dimension complementary characteristics of runoff inflow between watersheds.

[0230] 2.2.1 Earnings Differences

[0231] This implementation plan proposes to adopt a runoff complementarity ratio. Assess the surplus differences between different watersheds. Runoff complementarity rate. It uses multi-year averages as a reference standard, based on the specific conditions of each river basin. Runoff volume over a specific time period is used to determine the feasibility of runoff complementarity between watersheds, and the complementarity rate. This represents the percentage of complementary runoff in each time period. Taking watersheds A and B as examples, the calculation formula is as follows:

[0232] ;

[0233] ;

[0234] ;

[0235] ;

[0236] In the formula, This represents the runoff complementarity rate, which measures the proportion of the total amount of water that can theoretically compensate for each other between two watersheds over the entire scheduling period (T time periods) to the sum of the total runoff of the two watersheds. The larger the k value, the greater the potential for the surplus water of one watershed to effectively compensate for the water shortage in the other watershed, that is, the better the surplus complementarity between the watersheds. This represents the compensable water volume for each time period from time period 1 to time period T. The sum of . This represents the potential dispatchable water volume between watershed A and watershed B during time period t. This represents the total runoff of watersheds A and B during the entire scheduling period (T time periods). and They represent time periods t and t respectively. , Watershed runoff; and express , The average annual runoff of the watershed; and express , River Basin The deviation between runoff during a given period and the multi-year average runoff. The magnitude of the value reflects the complementary flow rate. and The proportion of the total runoff of the two streams The larger the value, the better the runoff complementarity between watersheds; conversely, the smaller the value, the worse the complementarity.

[0237] 1.2.2 Volatility Differences

[0238] The runoff complementarity coefficient is used to quantify the differences in runoff fluctuations between watersheds, reflecting the degree of runoff complementarity between watersheds throughout the entire scheduling period. The calculation formula is as follows:

[0239] ;

[0240] ;

[0241] In the formula, This represents the runoff complementarity coefficient between watersheds. This coefficient is used to comprehensively measure the degree to which the changes in runoff between watersheds cancel each other out in a system composed of multiple watersheds over adjacent time periods. The closer the value is to 0, the stronger the complementarity of runoff fluctuations in each basin within the system. That is, the increase in one basin can be well offset by the decrease in another basin, and the more stable the total inflow of the entire system. The larger the value, the more synchronized the runoff fluctuations in each basin tend to be, the weaker the complementarity, and the more significant the fluctuations in the total inflow of the system. Indicates the total number of time periods; Indicates the calculation period (ten-day period, month, quarter); Indicates the runoff number The degree of complementarity of runoff fluctuations during each change; Indicates the number of watersheds; Represents a watershed Flow in the first The change in the k+1th time period (runoff in the kth time period - runoff in the kth time period); runoff complementarity coefficient Take all calculation periods The average value.

[0242] 1.3 Comprehensive Complementary Characteristics Indicators

[0243] The above sub-indices are weighted and averaged to form a spatiotemporal complementarity composite index. This index can directly quantify the strength of cross-basin runoff complementarity.

[0244] 1. Calculate the values ​​of each indicator:

[0245] Connectivity index, runoff consistency index, runoff complementarity rate, and runoff complementarity coefficient.

[0246] 2. Determine the objective weights of each indicator using the entropy weight method:

[0247] The core logic of the entropy weight method is to reflect the degree of data dispersion through the entropy value of the indicator. The higher the dispersion, the more effective information the indicator provides, and the greater the corresponding weight. The specific calculation involves three steps:

[0248] (1) Standardization

[0249] Standardization of positive indicators (higher values ​​are better, such as runoff consistency index and runoff complementarity rate):

[0250] ;

[0251] Standardization of negative indicators (lower values ​​are better, such as connectivity indicators and runoff complementarity coefficients):

[0252] ;

[0253] in, Indicates the first A sample (e.g., different years, different regions) in the th... The raw data of each indicator (consistency between abundant and scarce water levels, runoff complementarity rate, etc.); Indicates the first The maximum value of each indicator among all samples; Indicates the first The minimum value of each indicator across all samples. Specifically, The subscript i represents the i-th sample. A sample refers to the specific object participating in the evaluation. For example, in different application scenarios, a sample can refer to a certain year, a specific pair of watersheds, or a specific region. If analyzing data from 10 years and 4 indicators, i ranges from 1 to 10. The subscript j represents the j-th indicator. An indicator refers to different dimensions or characteristics used for evaluation. In this invention, it refers to the four core indicators calculated: the consistency indicator between abundance and scarcity, the consistency indicator of changing trends, the surplus difference indicator, and the volatility difference indicator. If analyzing these 4 indicators simultaneously, j ranges from 1 to 4.

[0254] The positive indicator standardization formula linearly projects the original data to the [0, 1] interval. The worst sample receives a score of 0 after standardization, and the best sample receives a score of 1. This ensures that the standardized value is positively correlated with the indicator's superiority. The negative indicator standardization formula also projects the data to the [0, 1] interval, but performs the reverse processing. The worst sample receives a score of 0 after standardization, and the best sample receives a score of 1.

[0255] (2) Calculate the information entropy of each indicator. According to the definition of information entropy in information theory, the information of a set of data is calculated as follows:

[0256] ;

[0257] in:

[0258] ;

[0259] In the formula, Let be the information entropy of the j-th indicator, where the subscript j represents the j-th evaluation indicator. In the entropy weight method, The smaller the value, the greater the data variation among samples under that indicator, meaning the higher the data dispersion and the greater the amount of information it provides. Therefore, it should be given greater weight in the overall evaluation. Conversely, The larger the value, the more uniform the data, the less information it contains, and the smaller the weight should be. It can be expressed as the probability proportion of the value of the i-th sample under the j-th indicator, where the subscript i represents the i-th sample and the subscript j represents the j-th indicator. It can be represented as the value of the i-th sample on the j-th index after standardization. n is the total number of samples, that is, the number of all samples (different years, different watershed pairs) participating in the evaluation.

[0260] (3) Determine the weight of each indicator. Based on the formula for calculating information entropy, calculate the information entropy of each indicator as follows: The weights of each indicator are calculated using information entropy:

[0261] ;

[0262] In the formula, It is the calculated weight of the i-th indicator, which represents the importance of that indicator in the evaluation system composed of all indicators. The information entropy of the i-th indicator. In the entropy weight method, the greater the dispersion of the data sequence of an indicator (i.e., the more obvious the difference in the value of each sample on the indicator), the smaller its information entropy value, indicating that the indicator can provide more valuable information in the comprehensive evaluation and is more important. This represents the total number of indicators used in the comprehensive evaluation. 1- Let be the difference coefficient of the i-th indicator. Since a larger information entropy indicates a smaller information utility value, 1- The larger the value, the greater the degree of difference in the indicator, and the greater the weight it should be given in the evaluation.

[0263] 3. Calculate the comprehensive complementarity index ( ):

[0264] ;

[0265] in, Indicators representing spatiotemporal complementarity This indicates the number of complementarity evaluation indicators. Indicates the first The weight of each indicator, Indicates the first The value of each indicator is the standardized value of the i-th evaluation indicator. Standardization is to eliminate the influence of different indicator dimensions, making them dimensionless and comparable values.

[0266] 4. Grading of Complementarity Strength. When conducting complementarity strength analysis, relevant indicators should be used to quantify the strength of runoff complementarity. Therefore, the following methods are employed: An index is used to classify runoff inflows, and its expression is:

[0267] ;

[0268] ;

[0269] ;

[0270] ;

[0271] in, Indicates the first The Z-index for each time period is used to quantify the complementarity level of runoff inflow. By comparing it with the theoretical quantile of the standard normal distribution (e.g., Z=1.645 corresponds to the 95th quantile), the complementarity can be objectively classified into levels such as very strong, strong, average, weak, and very weak. It represents the skewness coefficient, reflecting the degree of asymmetry in the distribution of runoff series. Indicates the first Standardized variables for runoff over a period of time are used to eliminate the dimensional effects of runoff. Indicates the first Measured runoff values ​​for each time period. The arithmetic mean of the runoff series. This indicates the number of samples in the runoff series. The coefficient for calculating the skewness coefficient. The standard deviation represents the runoff series and reflects the degree of dispersion of the runoff.

[0272] Table 2. Z-index Classification Standards

[0273]

[0274] Z = ±1.645 corresponds to a 5% probability (i.e., a 90% confidence interval) on both sides of a standard normal distribution; Z = ±0.842 corresponds to a 20% probability (i.e., a 60% confidence interval) on both sides of a standard normal distribution. Figure 3 As shown.

[0275] This implementation plan constructs a complete and quantifiable indicator system, transforming the natural phenomenon of "spatiotemporal complementarity of runoff" into a precise mathematical representation. This provides an objective standard for judging "what constitutes 'strong complementarity'," significantly improving the scientific nature of decision-making. Furthermore, this invention is the first to systematically analyze cross-basin runoff complementarity from two dimensions: "consistency" and "difference." It not only focuses on the runoff characteristics of the basin itself but also emphasizes the interaction relationships between basins (such as encounters between abundant and scarce periods, and fluctuation offsetting), providing a systematic tool for comprehensively assessing cross-basin complementarity potential.

[0276] According to an embodiment of the present invention, a cross-basin hierarchical scheduling device based on runoff spatiotemporal complementarity assessment is provided, comprising:

[0277] The acquisition module is used to acquire runoff time series data of multiple watersheds within a preset scheduling period;

[0278] The calculation module is used to calculate time dimension indicators and spatial dimension indicators based on the runoff time series data; wherein, the time dimension indicators are used to assess the temporal synchronicity of runoff changes between watersheds, and the spatial dimension indicators are used to assess the differences in runoff volume surplus and fluctuation between watersheds.

[0279] The determination module is used to determine the objective weights corresponding to each indicator based on the numerical distribution of the time dimension indicators and the spatial dimension indicators, according to a preset entropy weight method.

[0280] The generation module is used to generate a spatiotemporal comprehensive complementarity index by weighted summation of the standardized index values ​​according to the objective weights.

[0281] A conversion module is used to convert the spatiotemporal comprehensive complementarity index into a Z-index that follows a standard normal distribution;

[0282] The output module is used to determine the runoff complementarity level among the multiple watersheds based on the quantile interval of the Z-index under the standard normal distribution.

[0283] The generation module is used to generate scheduling control signals for inter-basin water transfer projects based on runoff complementarity levels.

[0284] According to an embodiment of the present invention, an electronic device is provided; please refer to... Figure 4 The electronic device in this embodiment may include one or more of the following components: a processor, a network interface, memory, non-volatile memory, and one or more application programs, wherein the one or more application programs may be stored in non-volatile memory and configured to be executed by one or more processors, and the one or more programs are configured to perform the methods as described in the foregoing method embodiments.

[0285] According to embodiments of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a computer, causes the computer to perform the method described in any of the above embodiments.

[0286] According to embodiments of the present invention, a computer program product comprising instructions is also provided, which, when executed by a computer, cause the computer to perform a method in any of the above embodiments.

[0287] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A cross-basin hierarchical scheduling method based on runoff spatiotemporal complementarity assessment, characterized in that, The method includes: Acquire runoff time series data from multiple watersheds within a preset scheduling period; Based on the runoff time series data, time dimension indicators and spatial dimension indicators are calculated; wherein, the time dimension indicators are used to assess the temporal synchronicity of runoff changes between watersheds, and the spatial dimension indicators are used to assess the differences in runoff volume surplus and deficit and fluctuation between watersheds. Based on the preset entropy weight method, the objective weights corresponding to each indicator are determined according to the numerical distribution of the time dimension indicators and the spatial dimension indicators. The standardized index values ​​are weighted and summed according to the objective weights to generate a spatiotemporal comprehensive complementarity index. The spatiotemporal comprehensive complementarity index is converted into a Z-index that follows a standard normal distribution; Based on the quantile interval of the Z-index under the standard normal distribution, the runoff complementarity level among multiple watersheds is determined. Based on the runoff complementarity level, scheduling control signals are generated for controlling inter-basin water transfer projects.

2. The method according to claim 1, characterized in that, The time dimension indicators include a first time dimension indicator and a second time dimension indicator; wherein, the first time dimension indicator is a consistency indicator for assessing the consistency of runoff abundance and scarcity between watersheds, and the second time dimension indicator is a consistency indicator for assessing the consistency of runoff change trends between watersheds. The spatial dimension indicators include a first spatial dimension indicator and a second spatial dimension indicator; wherein, the first spatial dimension indicator is a surplus difference indicator used to assess the runoff surplus compensation capacity between watersheds, and the second spatial dimension indicator is a fluctuation difference indicator used to assess the degree of runoff fluctuation offsetting between watersheds.

3. The method according to claim 2, characterized in that, The steps of calculating time dimension indicators and spatial dimension indicators based on the runoff time series data include: the steps of calculating the first time dimension indicator and the steps of calculating the second time dimension indicator; The steps for calculating the first time dimension index include: Based on a preset runoff cumulative frequency threshold, the runoff data for each time period are divided into multiple abundant and scarce levels. Based on the abundance and scarcity levels of runoff sequences in two watersheds at various time periods, the abundance and scarcity consistency index is calculated according to the number of time periods with the same level, the number of time periods with a level difference of the first preset level, the number of time periods with a level difference of the second preset level, and the number of time periods with opposing levels. The steps for calculating the second time dimension index include: Calculate the rate of change of the runoff time series data in adjacent time periods; Based on the difference in the runoff change rate between the two watersheds at corresponding time periods, a consistency index of change trend is calculated.

4. The method according to claim 3, characterized in that, The step of calculating time dimension indicators and spatial dimension indicators based on the runoff time series data includes: the step of calculating the first spatial dimension indicator and the step of calculating the second spatial dimension indicator; The step of calculating the first spatial dimension index includes: Based on the preset multi-year average runoff of each watershed, the runoff deviation value in each time period is calculated. For each time period, if the runoff deviation values ​​of the two watersheds have opposite signs, the compensable water volume for that time period is determined based on the absolute value of the runoff deviation values ​​of the two watersheds. The surplus difference index is calculated based on the ratio of the sum of compensable water volume for all time periods during the scheduling period to the sum of the total runoff of the two watersheds. The steps for calculating the second spatial dimension index include: Calculate the runoff changes in the multiple watersheds during each time period within the scheduling period; The fluctuation difference index is calculated based on the algebraic sum of the runoff changes in the multiple watersheds during the same period.

5. The method according to claim 4, characterized in that, The step of determining the objective weights corresponding to each indicator based on the preset entropy weight method and the numerical distribution of the time dimension indicators and the spatial dimension indicators includes: The consistency index between abundant and scarce periods, the consistency index of changing trends, the surplus difference index, and the volatility difference index are standardized; wherein, the surplus difference index is standardized using a positive standardization method, and the consistency index between abundant and scarce periods and the volatility difference index are standardized using a negative standardization method. Based on the standardized indicator values, calculate the information entropy of each indicator; Calculate the corresponding objective weight based on the information entropy of each indicator; the greater the dispersion of the indicator, the smaller its information entropy, and the greater the weight assigned to it.

6. The method according to claim 1, characterized in that, The step of converting the spatiotemporal comprehensive complementarity index into a Z-index that follows a standard normal distribution includes: The skewness coefficient of the spatiotemporal comprehensive complementarity index is calculated to quantify the asymmetry of its distribution; Based on the skewness coefficient, the spatiotemporal comprehensive complementarity index is normalized to generate a Z-index that follows or approximately follows a standard normal distribution; wherein, the Z-index is used to eliminate the influence of the spatiotemporal comprehensive complementarity index on the complementarity level determination caused by the difference in the original distribution form.

7. The method according to claim 6, characterized in that, The step of determining the runoff complementarity level among multiple watersheds based on the quantile interval of the Z-index under the standard normal distribution includes: The Z-index is compared with a plurality of preset thresholds; wherein the plurality of thresholds include a first preset threshold, a second preset threshold, a third preset threshold and a fourth preset threshold whose values ​​decrease sequentially, and the thresholds are determined based on the quantiles of a standard normal distribution. Based on the comparison results, the runoff complementarity levels are divided into five levels: extremely strong complementarity, strong complementarity, average complementarity, weak complementarity, and extremely weak complementarity; among them... When the value of the Z-index is greater than or equal to the first preset threshold, the complementarity level is determined to be extremely strong complementarity. When the value of the Z-index is less than the first preset threshold and greater than or equal to the second preset threshold, the complementarity level is determined to be strong complementarity. When the value of the Z-index is less than the second preset threshold and greater than or equal to the third preset threshold, the complementarity level is determined to be flat complementarity. When the value of the Z-index is less than the third preset threshold and greater than or equal to the fourth preset threshold, the complementarity level is determined to be weak complementarity. When the value of the Z-index is less than the fourth preset threshold, the complementarity level is determined to be very weak complementarity.

8. A cross-basin hierarchical scheduling device based on runoff spatiotemporal complementary characteristics assessment, characterized in that, include: The acquisition module is used to acquire runoff time series data of multiple watersheds within a preset scheduling period; The calculation module is used to calculate time dimension indicators and spatial dimension indicators based on the runoff time series data; wherein, the time dimension indicators are used to assess the temporal synchronicity of runoff changes between watersheds, and the spatial dimension indicators are used to assess the differences in runoff volume surplus and fluctuation between watersheds. The determination module is used to determine the objective weights corresponding to each indicator based on the numerical distribution of the time dimension indicators and the spatial dimension indicators, according to a preset entropy weight method. The generation module is used to generate a spatiotemporal comprehensive complementarity index by weighted summation of the standardized index values ​​according to the objective weights. A conversion module is used to convert the spatiotemporal comprehensive complementarity index into a Z-index that follows a standard normal distribution; The output module is used to determine the runoff complementarity level between multiple watersheds based on the quantile interval of the Z-index under the standard normal distribution. The generation module is used to generate scheduling control signals for inter-basin water transfer projects based on runoff complementarity levels.

9. An electronic device, characterized in that, include: A memory, and one or more processors communicatively connected to the memory; The memory stores instructions that can be executed by the one or more processors to cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images